The utility of texture analysis to improve per-pixel classification for high to very high spatial resolution imagery

Earth observation data are becoming available at increasingly finer resolutions. Sensors already in existence (IKONOS, Quickbird. SPOT 5. Orbview) or due to be launched in the near future will reach 1-5 tn resolution. These very high resolution (VHR) data will provide more details ofthe urban areas, but it seems evident that they will create additiotial problems in terms oi' infonnation extraction using automatic classification, ln this fi'amework. this paper examines the potential of the spectral/textural approach to improve the classification accuracy of intra-urban land eover types. The utility ofthe textura! analysis was measured in comparison with multi-speclral per-pixel classifications. Haralick's second-order statistics were applied to the co-occurrence matrix. Four texture indices with six window sizes created from panchromatic images were tested on images at high to very high re.solulions (10-1 m). The results show that the optimal index improving ihe global classification accuracy is the homogeneity measure, with ul xl window size. Moreover, for 1 m images. tcxUire measure of homogeneity allows one to decrease the shadows.